基于经验模式分解的直觉模糊网络故障诊断  

Fault Diagnosis with Empirical Model Decomposition and Intuitionistic Fuzzy Reasoning Neural Networks

作　　者：许翔宇[1] 黄席樾[1] 赵勇[1] 黄勇[1] 

机构地区：[1]重庆大学自动化学院,重庆400044

出　　处：《重庆理工大学学报（自然科学）》2010年第4期91-96,共6页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然基金资助项目(50975011)

摘　　要：针对平稳时间序列信号,提出一种基于EMD多模态分量特征提取和直觉模糊网络的故障诊断方法。该方法首先对原始信号进行EMD分解,经EMD分解获得基本模式分量(IMF),选择能量最大的几个IMF并转化为模糊特征向量,对机器故障进行诊断,然后将模糊特征向量输入到直觉模糊网络中,实现对机器工作状态及不同故障类型的识别,该方法应用于柴油机振动信号的故障诊断,实验结果证明了其可行性和有效性。A novel fault diagnosis method based on Empirical Model Decomposition and Intuitionistic Fuzzy Reasoning Neural Networks is proposed to deal with steady time sequence singals in this paper.Firstly,the mechanical fault can be diagnosed by decomposing sampling-signals with EMD to obtain intrinsic mode functions（IMFs）,and then selecting several dominating energy IMFs to be converted into fuzzy feature vectors.Secondly,the extracted fuzzy feature vectors are input into the Intuitionistic Fuzzy Reasoning Neural Networks to detect the different abnormal cases,The application in fault diagnosis of diesel engine shows feasible and efficient performance of the proposed method.

关 键 词：故障诊断 经验模态分解 基本模式分量 直觉模糊网络 

分 类 号：TP207[自动化与计算机技术—检测技术与自动化装置]